The long term goal of this proposal is to understand how receptive fields and functional maps in the primary visual cortex are wired during development. Over the years, a number of theoretical models have been put forward to explain the origin of the columnar organization of visual cortex. Despite numerous experimental and theoretical advances, key questions remain open about the function and source of columnar systems. These include the actual origin of functional maps, their mutual relationships and, most importantly, their significance for normal visual processing. Our working hypothesis is that 'statistical connectivity'between the thalamus and the cortex provides an initial blueprint for the formation columnar maps that is subsequently maintained, and modified, by activity- dependent processes during the critical period. We propose to elaborate, generalize, and critically test this idea by addressing the following aims. Aim 1: We will quantify various features of the predicted maps, their mutual relationships, and compare the results to experimental data. We will also investigate how the local anisotropy of the retinotopic map is related to the structure of other feature maps. Aim 2: We will investigate how neuronal selectivity depends on the location of the cell across the functional maps. Aim 3: We will study how existing activity-dependent developmental models could operate to maintain, or modify, the initial structures set up by statistical connectivity. Aim 4: We will test a novel hypothesis regarding the formation of ocular dominance columns that overcomes the initial dominance by contralateral input. The proposed work will critically evaluate the merits and weaknesses of statistical connectivity theory in explaining a myriad of published experimental findings and generate novel predictions. If the theory stands up to the test of time, it could potentially provide a unified explanation for the origin of the cortical architecture and clarify which of its features are expected to be modified by activity-dependent mechanisms during development. Altogether, these studies could radically alter the way we look at the wiring of receptive fields and maps in early visual cortex and the functional significance of cortical maps for vision.